People aren't great at judging risks. Example: Lichtenstein et al. (1978) Judged Frequency of Lethal Events Hence, we should not rely on our "gut", better to use numbers! De Veaux and Velleman (2009) Math is Music, Statistics is Literature : The idea here: it takes a while, and experience, to develop the skills/thinking necessary to be good at statistics. It is more common to find mathematical prodigies. (As to the analogy between math and music, that's a shade more iffy. At the very least, I think this refers mostly to Western classical music. Hindustani music/jazz has fewer prodigies -- musicians tend to reach their peaks relatively late in life in these forms of music.) Daniel Bernoulli's work in the 1800s on maximizing expected utility: Daniel Ellsberg: Ellsberg's paradox (1961) suggests that people are averse to ambiguity. No utility function is consistent wit h the choice that people make.